Optimum Design and Estimation of Step‐Stress Accelerated Life Testing for Chen Distribution Under an Adaptive Type‐II Progressive Hybrid Censored Data

Abstract

ABSTRACT This paper explores the significant advantages of using adaptive Type‐II progressive hybrid censoring in step‐stress accelerated life tests. By implementing this method, researchers can significantly reduce testing time and costs while improving the precision of statistical analysis. The study focuses on ‐level step‐stress accelerated life testing, incorporating adaptive Type‐II progressive hybrid censoring and assuming that test unit lifetimes follow the Chen distribution. In this model, the scale parameter of the Chen failure time distribution is treated as a log‐linear function of the stress level. The paper develops both maximum likelihood and Bayesian estimators for key model parameters, providing estimated confidence intervals through an approximate normal method based on asymptotic distributions. Additionally, an optimal testing strategy is introduced to determine the ideal timing for stress level increases, minimizing the and optimality of the MLE of the model parameters. The paper concludes with a numerical data analysis, providing concrete examples and applications for these concepts.